Reciprocal associations between implicit attitudes and drinking in emerging adulthood

Abstract

Background: Implicit alcohol attitudes are considered important in the etiology of drinking, and theory posits reciprocal associations between them. Research testing reciprocal associations between implicit attitudes (using the Implicit Association Task, IAT) and drinking is limited by a failure to consider multiple processes influencing performance on the IAT and to disaggregate within- and between-person effects. The current study addressed these limitations by using a diffusion model to analyze IAT data and Latent Curve Models with Structured Residuals to test reciprocal associations.

Methods: The sample included 314 emerging adults from the community (52% female; predominantly non-Hispanic Caucasian (76%) or African American (15%)) assessed annually for three years. Differences between IAT conditions in the drift rate parameter of the EZ-diffusion model (vΔ) were used as an alternative to traditional response-time-based indices from the IAT (d-scores). Differences in drift rate have been found to index implicit attitudes effectively.

Results: Within-person reciprocal associations were supported, but between-person associations were not. Positive implicit alcohol attitudes (vΔ) were prospectively associated with heavy drinking, which was positively associated with subsequent positive implicit alcohol attitudes.

Conclusions: We found that positive implicit alcohol attitudes and heavy drinking reinforce each other in a negative cascade within individuals. The results highlight the importance of disaggregating within- and between-person prospective effects when testing dual process models and suggest that the diffusion model may be a fruitful approach to enhance the construct validity of IAT assessed implicit attitudes.

Keywords: dual process theory; emerging adulthood; heavy drinking; implicit attitudes; response time measures.

Figure 1.

Schematic of the decision process assumed by the EZ diffusion model for a “compatible” condition SC-IAT trial in which an alcohol stimulus is presented. Note. v = drift rate, or efficiency with which an individual gathers evidence in favor of the correct choice. a = boundary separation; response conservativeness (e.g., speed/accuracy trade-off settings). Ter = nondecision time (e.g., motor response speed). The model assumes that, on a given trial, an evidence accumulation process drifts between a boundary for the correct choice, set at parameter a, and a boundary for the incorrect choice, set at 0. The process begins at a start point (in the simplified EZ diffusion model framework, always assumed to be a/2) and drifts toward the correct choice boundary with an average rate of v. The Ter parameter is a constant that accounts for the time taken up by other processes peripheral to the decision process (e.g., perceptual encoding, motor responses). The evidence accumulation process typically terminates at the upper, correct choice boundary (for this trial, the “alcohol”/“good” choice boundary), but errors occur when noise causes the process to terminate at the lower boundary (for this trial, the “bad” choice boundary). Gray traces represent simulated decision processes on individual trials. Gray density plots represent the density of response times at the respective boundaries that are predicted by the model.

Figure 2.

Latent Curve Model with Structured Residuals for Implicit Attitudes and Heavy Drinking. Note. Solid black lines are significant and dotted grey lines are non-significant pathways. Betas are reported next to significant associations and standard errors are reported in parentheses. Levels of significance were based on unstandardized regression estimates. For simplicity, the covariates of between-condition changes in nondecision time and between-condition changes in response conservativeness across Waves 7–9 are not depicted. RI = Random Intercept. W = Wave. p < .05 = * p < .01 = ** p < .001 = ***.